-
Notifications
You must be signed in to change notification settings - Fork 36
Expand file tree
/
Copy pathndarray-eigen.cc
More file actions
152 lines (136 loc) · 5.25 KB
/
Copy pathndarray-eigen.cc
File metadata and controls
152 lines (136 loc) · 5.25 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
// -*- c++ -*-
/*
* Copyright (c) 2010-2012, Jim Bosch
* All rights reserved.
*
* ndarray is distributed under a simple BSD-like license;
* see the LICENSE file that should be present in the root
* of the source distribution, or alternately available at:
* https://github.com/ndarray/ndarray
*/
#include "ndarray/eigen.h"
#include "Eigen/SVD"
#define BOOST_TEST_DYN_LINK
#define BOOST_TEST_MODULE ndarray-eigen
#include "boost/test/unit_test.hpp"
template <typename T, typename U>
void testElements2(T const & a, U const & b) {
BOOST_CHECK( a.template getSize<0>() == b.rows() );
BOOST_CHECK( a.template getSize<1>() == b.cols() );
BOOST_CHECK( a.template getStride<0>() == b.rowStride() );
BOOST_CHECK( a.template getStride<1>() == b.colStride() );
for (int i = 0; i < b.rows(); ++i) {
for (int j = 0; j < b.cols(); ++j) {
BOOST_CHECK(&a[i][j] == &b.coeffRef(i,j));
}
}
}
template <typename T, typename U>
void testElements1(T const & a, U const & b) {
BOOST_CHECK( a.template getSize<0>() == b.size() );
BOOST_CHECK( a.template getStride<0>() == b.innerStride() );
for (int i = 0; i < b.size(); ++i) {
BOOST_CHECK(&a[i] == &b.coeffRef(i));
}
}
template <typename T, int N, int C>
ndarray::Array<T, N, C> makeMutable(ndarray::Array<T, N, C> const & a) {
return a;
}
template <typename T, int N, int C>
ndarray::ArrayRef<T, N, C> makeMutable(ndarray::ArrayRef<T, N, C> const & a) {
return a;
}
template <typename T, int N, int C>
ndarray::Array<T, N, C> makeMutable(ndarray::Array<T const, N, C> const & a) {
return ndarray::const_array_cast<T>(a);
}
template <typename T, int N, int C>
ndarray::ArrayRef<T, N, C> makeMutable(ndarray::ArrayRef<T const, N, C> const & a) {
return ndarray::const_array_cast<T>(a.shallow()).deep();
}
template <int Rows, typename XprKind, typename A>
void testAsEigen1(A const & a) {
typedef typename boost::remove_const<typename A::Element>::type T;
ndarray::asEigen<XprKind>(makeMutable(a)).setRandom();
int const n = asEigenMatrix(a).size();
testElements1(a, ndarray::asEigen<XprKind>(a));
Eigen::Matrix<T, Rows, Eigen::Dynamic> m1(n, 6);
m1.setRandom(n, 6);
Eigen::Matrix<T, Eigen::Dynamic, 1> m2(6, 1);
m2.setRandom(6, 1);
asEigenMatrix(makeMutable(a)) = m1 * m2;
Eigen::Matrix<T, Rows, 1> m3 = m1 * m2;
for (int i = 0; i < n; ++i) {
BOOST_CHECK(a[i] == m3(i));
}
Eigen::Array<T, Rows, 1> m4(n, 1);
m4.setRandom();
Eigen::Array<T, Rows, 1> m5 = m4 * asEigenArray(a);
Eigen::Array<T, Rows, 1> m6 = m4 * m3.array();
BOOST_CHECK((m5 == m6).all());
}
template <int Rows, int Cols, typename XprKind, typename A>
void testAsEigen2(A const & a) {
typedef typename boost::remove_const<typename A::Element>::type T;
ndarray::asEigen<XprKind>(makeMutable(a)).setRandom();
int const m = asEigenMatrix(a).rows();
int const n = asEigenMatrix(a).cols();
testElements2(a, ndarray::asEigen<XprKind>(a));
Eigen::Matrix<T, Rows, Eigen::Dynamic> m1(m, 6);
m1.setRandom(m, 6);
Eigen::Matrix<T, Eigen::Dynamic, Cols> m2(6, n);
m2.setRandom(6, n);
asEigenMatrix(makeMutable(a)) = m1 * m2;
Eigen::Matrix<T, Rows, Cols> m3 = m1 * m2;
for (int i = 0; i < m; ++i) {
for (int j = 0; j < n; ++j) {
BOOST_CHECK(a[i][j] == m3(i, j));
}
}
Eigen::Array<T, Rows, Cols> m4(m, n);
m4.setRandom();
Eigen::Array<T, Rows, Cols> m5 = m4 * asEigenArray(a);
Eigen::Array<T, Rows, Cols> m6 = m4 * m3.array();
BOOST_CHECK((m5 == m6).all());
}
template <typename T, typename XprKind>
void invokeAsEigenTests() {
ndarray::Array<T, 2, 2> a22(ndarray::allocate(5, 4));
testAsEigen2<5, 4, XprKind>(a22);
ndarray::Array<T, 2, 1> a21(a22[ndarray::view()(0, 3)]);
testAsEigen2<5, 3, XprKind>(a21);
testAsEigen2<3, 5, XprKind>(a21.transpose());
ndarray::Array<T, 2, 0> a20(a22[ndarray::view()(0, 4, 2)]);
testAsEigen2<5, 2, XprKind>(a20);
testAsEigen2<2, 5, XprKind>(a20.transpose());
ndarray::Array<T, 1, 1> a11(ndarray::allocate(4));
testAsEigen1<4, XprKind>(a11);
testAsEigen1<4, XprKind>(a11.transpose());
ndarray::Array<T, 1, 0> a10(a11[ndarray::view(0, 4, 2)]);
testAsEigen1<2, XprKind>(a10);
testAsEigen1<2, XprKind>(a10.transpose());
}
BOOST_AUTO_TEST_CASE(AsEigen) {
invokeAsEigenTests<double, Eigen::ArrayXpr>();
invokeAsEigenTests<double, Eigen::MatrixXpr>();
invokeAsEigenTests<float const, Eigen::ArrayXpr>();
invokeAsEigenTests<float const, Eigen::MatrixXpr>();
}
template <typename SVD, typename Matrix, typename Vector>
void testSVD(Matrix const & a, Vector const & b, Vector & x) {
SVD svd(a, Eigen::ComputeThinU | Eigen::ComputeThinV);
x = svd.solve(b);
BOOST_CHECK((a.transpose() * a * x).isApprox(a.transpose() * b));
}
BOOST_AUTO_TEST_CASE(SVD) {
auto aArray = ndarray::Array<double, 2, 2>(ndarray::allocate(8,5));
auto bArray = ndarray::Array<double, 1, 1>(ndarray::allocate(8));
auto xArray = ndarray::Array<double, 1, 1>(ndarray::allocate(5));
auto a = ndarray::asEigenMatrix(aArray);
auto b = ndarray::asEigenMatrix(bArray);
auto x = ndarray::asEigenMatrix(xArray);
a.setRandom();
b.setRandom();
testSVD<Eigen::JacobiSVD<Eigen::MatrixXd>>(a, b, x);
}